Multi-uav cooperative bridge bottom disease detection and three-dimensional positioning method and system

The bridge underpass defect detection system, which utilizes multi-UAV collaborative technology, solves the problems of positioning blind spots and duplicate defect statistics in environments without GNSS signals, and achieves high-precision three-dimensional positioning and full-coverage detection of bridge defects by employing UWB positioning and multi-view collaborative technology.

CN122415561APending Publication Date: 2026-07-17XINJIANG UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
XINJIANG UNIVERSITY
Filing Date
2026-04-28
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing UAV-based bridge underpass detection methods suffer from positioning failures in environments without GNSS signals, visual blind spots, and duplicate statistics of defects, making it difficult to achieve high-precision three-dimensional positioning and all-round coverage.

Method used

A multi-UAV collaborative approach is adopted, utilizing UWB local positioning network and time-of-flight algorithm to obtain three-dimensional spatial coordinates. Combined with multi-view collaborative formation flight and YOLOv8 target detection network to identify disease targets, the DeepSORT algorithm is used to perform cross-frame and cross-view data association to generate three-dimensional positioning data of the disease.

Benefits of technology

It achieves high-precision positioning and all-round coverage in complex bridge under-bridge environments, eliminates visual blind spots, avoids repeated counting of defects, and provides accurate data support for bridge health assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122415561A_ABST
    Figure CN122415561A_ABST
Patent Text Reader

Abstract

本发明公开了一种多无人机协同的桥底病害检测与三维定位方法及系统。针对桥底全球导航卫星系统(GNSS)拒止环境,该方法部署超宽带UWB定位网络实时解算机载三维坐标;结合领航‑跟随者模型与人工势场法实现多机编队协同控制,引入硬件触发机制完成图像与位姿的微秒级时空同步采集;采用改进的YOLOv8网络进行病害目标检测以输出病害二维边界框,基于DeepSORT算法融合马氏与外观余弦距离构建综合代价矩阵,实现跨视角关联去重及全局唯一标识分配;融合同步位姿与桥底刚体变换模型,对去重二维框反投影生成病害三维绝对坐标。本发明克服了信号遮挡与视觉盲区,实现了交叠视野下的病害精准去重与高精度三维溯源定位。
Need to check novelty before this filing date? Find Prior Art